Related Experiment Video
Updated: Sep 9, 2025
13:19
Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
9.3K
Leveraging Deep Learning to Enhance Malnutrition Detection via Nutrition Risk Screening 2002: Insights from a
Nadir Yalçın1, Merve Kaşıkcı2, Burcu Kelleci-Çakır1
1Department of Clinical Pharmacy, Faculty of Pharmacy, Hacettepe University, Ankara 06100, Türkiye.
Nutrients
|August 28, 2025
Summary
A new machine learning tool accurately predicts the need for and type of nutritional therapy. Integrating demographic data with the Nutrition Risk Screening 2002 (NRS-2002) enhances prediction accuracy for enteral, parenteral, or combined nutrition.
Area of Science:
- Clinical Nutrition
- Artificial Intelligence in Healthcare
- Medical Informatics
Background:
- Accurate nutritional assessment is crucial for patient outcomes.
- Existing tools like Nutrition Risk Screening 2002 (NRS-2002) have limitations in predicting specific nutritional therapy needs.
- Machine learning (ML) offers potential for enhanced predictive capabilities in clinical decision-making.
Purpose of the Study:
- To develop and validate a novel ML-based screening tool for predicting the requirement and type (enteral, parenteral, combined) of nutritional therapy.
- To integrate the NRS-2002 with demographic parameters (gender, BMI, cancer status, hospital unit) for improved prediction.
- To utilize the Optimal Nutrition Care for All (ONCA) national cohort data for model development and validation.
Main Methods:
- A multicenter retrospective cohort study involving 191,028 patients.
- Development of a two-step ML classification model using Random Forest, Artificial Neural Network, and deep learning algorithms.
- Performance evaluation using metrics including Area Under the Curve (AUC), accuracy, sensitivity, and specificity.
Main Results:
- Deep learning (DL) demonstrated superior performance in both predicting the need for and type of nutritional therapy.
- Key predictors for nutritional therapy need included severe illness and reduced dietary intake (AUC=0.933).
- Important factors for therapy type prediction were severe illness, impaired nutritional status, and ICU admission (AUC=0.741); adding demographic data improved AUC by up to 3.27%.
Conclusions:
- An ML-enhanced model integrating NRS-2002 with demographic factors provides accurate classification of nutritional therapy needs and types.
- This tool can serve as a clinical decision support system to guide personalized nutritional therapy.
- Further external validation in larger, multinational cohorts is recommended before widespread clinical implementation.
More Related Videos
Related Concept Videos
02:35Key Elements for Plant Nutrition
21.2K
Like all living organisms, plants require organic and inorganic nutrients to survive, reproduce, grow and maintain homeostasis. To identify nutrients that are essential for plant functioning, researchers have leveraged a technique called hydroponics. In hydroponic culture systems, plants are grown—without soil—in water-based solutions containing nutrients. At least 17 nutrients have been identified as essential elements required by plants. Plants acquire these elements from the...
21.2K
01:27Parentral Nutrition: Centeral and Peripheral Parental Nutrition
300
Parenteral Nutrition (PN) delivers essential nutrients directly into the bloodstream, bypassing the digestive system. It is commonly used for individuals with severe digestive disorders or conditions that prevent normal nutrient absorption.
PN can be administered through two primary routes:
1. Central Parenteral Nutrition (CPN):
CPN involves delivering a high concentration of nutrients through a large vein. This is typically achieved using a Peripherally Inserted Central Catheter (PICC) or,...
PN can be administered through two primary routes:
1. Central Parenteral Nutrition (CPN):
CPN involves delivering a high concentration of nutrients through a large vein. This is typically achieved using a Peripherally Inserted Central Catheter (PICC) or,...
300